- Unset depth_cleared_level_mask for non-clear blits. Set the flag after
the clear, so that we don't have to check blitter_running.
- Set depth_cleared_level_mask only when we set depth_clear_value.
Fixes: ff8a930cf7 - radeonsi: add _once suffix to depth_cleared_level_mask
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13603>
si_update_ps_colorbuf0_slot used blitter_running as a way to detect
recursive calls.
Unfortunately this catch too many cases; for instance a backtrace
like:
#0 si_update_ps_colorbuf0_slot
#1 si_set_framebuffer_state
#2 do_blits
[...]
#5 si_blit
#6 si_copy_region_with_blit
Would end-up not updating ps_uses_fbfetch; so if the new fb_state is
something like:
cbufs = {0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0}, zsbuf = 0x55b8987545e0}
We can have ps_uses_fbfetch=true but cbufs[0] = NULL, which causes a
crash later in si_ps_key_update_framebuffer.
This commit fixes intermittent crashes in KHR-GL46.stencil_texturing.functional.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13550>
This eliminates the overhead of invoking si_decompress_depth.
The complication here is that we need to update needs_depth_decompress_mask
every time we update dirty_level_mask.
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13492>
ps is for the pixel shader, while ge is for VS, TCS, TES, and GS.
si_shader_key: 68 bytes
si_shader_key_ge: 68 bytes
si_shader_key_ps: 28 bytes
The only notable change is that si_shader_select_with_key is changed
to a C++ template. Other changes are trivial.
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13285>
It regresses the first snx test because it adds CPU overhead, and there is
no way to work around it. The average effect on viewperf is 0, meaning that
a few cases improve, while a few others regress.
Acked-by: Timur Kristóf <timur.kristof@gmail.com>
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13279>
SDMA support was dropped in 1f31a21664 mainly because the
advantages of delegating some copy/clear operations to the
SDMA hw came with large drawbacks: CPU overhead due to the
sdma/gfx synchronization and hangs.
This commit restores SDMA support for all gfx7+ chips but
only for the image copy operations.
SDMA operations won't be intertwined with gfx operations
like before. Instead, a SDMA IB will contain a single copy
at a time and the synchronization will be handled by the
winsys (based on the used buffers).
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12763>
This helps some viewperf subtests.
Only view XY culling is done. Edgeflags are always disabled with lines.
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13048>
Do it when we bind shaders.
The advantages are:
- no need to memset the fields when any shader variant state is changed
(e.g. culling on/off)
- no need to recompute the fields every time that happens
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12656>
We would like draw-only display lists to have immutable draw info and
this is the only GL non-draw state in pipe_draw_info (not counting
view_mask).
It also allows removing some code from draw_vbo for tessellation.
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12351>
When this flag is set, u_threaded_context will try not to map it directly
for better buffer placement. It's set by drivers when visible VRAM is too
small.
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12257>
The GDS ordered append variant is unstable due to kernel and firmware bugs.
The unordered GDS variant isn't faster than the memory-based variant.
Only the memory-based variant is kept.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11510>
Determine if a given shader write the same constant value to its output
if a specific input texture is replaced by constant load.
It's done by checking if the store_output intrinsics only depends on
constant and a texture. If it's true, the given texture is replaced by
a constant load in cloned shader and this clone is optimized.
Then the output is checked (= is it constant or not).
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10979>
This allows to implement custom draw_vbo code-path without
touching si_draw_vbo.
As an example, skipped all draw calls with an odd new_draws
could be done like this:
void mywrapper(...) {
if (new_draws % 2)
return;
return sctx->real_draw_vbo(...);
}
if (some_condition_is_met)
si_install_draw_wrapper(sctx, mywrapper);
Instead of having to add the "if ()" condition inside si_draw_vbo.
Note that a single wrapper may be installed so care must be taken
to not override an existing wrapper.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10979>